Page 18

Crawling robots The use of autonomous crawling robots also is under rapid development but has not reached the same commercial stage as flying drones (see figure 2 for an example of an experimental crawling robot). Many non-autonomous robots already are working in the solar and wind industries in a wide range of inspection tasks. One challenge for crawling robots is to “stick” to a vertical surface—like a wind turbine tower or the surface of a solar panel. Developers are experimenting with gecko-style nanotechnological adhesive materials, suction (using vacuum), and magnetic materials, among other technologies. An advantage of crawling robots over drones is that they can get close to a structure’s surface—and in fact touch it. This opens up possibilities for a new set of technologies, the most important being microwave and ultrasonic transmitters and receivers, which can be used to penetrate into the structure to reveal faults in materials. With crawling robots, artificial intelligence is used mainly to control them. But, as with flying drones, crawling robots collect large amounts of data, requiring artificial intelligence in the analysis of observations. Two obvious use-cases in renewables will be the autonomous inspection of wind towers and blades and around solar panels.

Figure 2 - 'Abigaille' wall-crawler robot (source: Simon Fraser University School of Engineering Science/MENRVA)

7 8

http://www.unmanned-ship.org/munin/ https://www.dnvgl.com/technology-innovation/revolt/

18 ENERGY Artificial intelligence

Driving robots Self-driving cars take the headlines, but for industry the arrival of self-driving trucks (as well as cranes and the like) may happen more quickly, due to labour cost savings. It is possible to imagine an onshore wind/solar farm being built entirely by autonomous robots: the parts of a wind tower and turbine or a solar array are transported from the factory by self-driving lorries, unloaded by another set of robots, attached to the foundations that yet other robots have dug and filled, and pieced together by a final set of robots and drones. Most of these robots are driving robots— or automatic guided vehicles—controlled by artificial intelligence. Artificial intelligence’s main use here pertains to obstacle avoidance and navigation, but higher level systems also are needed for the “orchestration” of the transport as well as actual construction. Still, though autonomous artificial intelligence-controlled robots in the transport and construction industry will continue to develop, the reality of an autonomously built renewable project is probably many years away. Sailing robots For the renewables industry, the sailing robot will transport and deliver parts, like the driving robots. And, like the driving robots, sailing robots need to use artificial intelligence to tackle the challenges of avoiding obstacles and navigating. Autonomous ships are still in the experimental stages (see figure 3 for an example). Several research projects (like MUNIN, which is looking at open-sea autonomous shipping7, and DNV GL’s ReVolt system, which is studying a coastal transportation system8) have investigated many aspects of autonomy. As with the other autonomous robots, the sensors used for giving the vehicle a picture of the surroundings are an essential part of these investigations.

Profile for DNV GL

Making renewables smarter  

The benefits, risks, and future of artificial intelligence in solar and wind industry - Artificial intelligence position paper

Making renewables smarter  

The benefits, risks, and future of artificial intelligence in solar and wind industry - Artificial intelligence position paper

Profile for dnvgl